Cell Culture Studies The antiretroviral efficacy of the NRTI AZT is dependent upon intracellular mono-, di- and tri-phosphorylation and incorporation of the drug into DNA in place of thymidine. Thymidine Kinase 1 (TK-1) catalyzes the first step of this pathway, and is considered to be critical to AZT efficacy as well as incorporation into DNA. We examined interindividual host variability in AZT metabolic capacity using 19 different strains of normal human mammary epithelial cells NHMECs established at reduction mammoplasty from 19 healthy women. In cells exposed for 24 hr to 200 M AZT, DNA was analyzed for AZT-DNA incorporation by AZT radioimmunoassay (RIA). Eight of the 19 NHMEC strains showed AZT-DNA incorporation levels of 16 to 50 molecules of AZT/106 nucleotides, 2 had 51 to 100 molecules of AZT/106 nucleotides, and 2 had &gt;100 molecules of AZT/106 nucleotides. The remaining 7 NHMEC strains did not have detectable AZT-DNA incorporation at 24 hr of exposure. In order to examine specific end points that might correlate with AZT-DNA incorporation, levels of the metabolically active (24KD) TK1 protein were determined in the 19 NHMEC strains by Western blot analysis. A statistically significant difference in TK-1 protein levels was observed among strains able to incorporate AZT into DNA at 24 hr and those that showed no measurable AZT-DNA incorporation at 24 hr (p= 0.0013, Mann-Whitney test). Overall, the data suggest that interindividual variability in AZT-DNA incorporation may be at least partially due to differences in the rate of TK-1 induction in response to AZT exposure. Novel mechanisms of genotoxicity, including centrosomal and kinetochore fragmentation have been found in hamster (CHO) cells and human (NHMECs) exposed to AZT for 24 hr. At mitosis the centrosome directs chromosomal migration, and normal kinetochore material is required for spindle formation. In CHO and NHMEC cells, dose-related AZT-induced centrosomal fragmentation was found by pericentrin staining, which revealed multipolar mitotic figures. In addition, loss of whole chromosomes from the nucleus was evidenced by kinetochore positive micronuclei. Abnormal tubulin distribution was also found in AZT-exposed cells containing &gt;2 centrosomes/cell. Immunofluorescent staining of AZT-exposed NHMECs with Aurora A, a kinase involved in the maturation of the centrosome, confirmed centrosomal amplification and multipolar mitotic figures that were not found in unexposed controls. Flow cytometry revealed similar cell cycle parameters in exposed and unexposed cells, confirming that cell division takes place in cells with &gt;2 centrosomes/nucleus. In order to extend these experiments to other NRTIs, CHO cells and NHMEC strains were exposed for 24 hr to 3TC, d4T, ddI or thymidine. In both cell types there was a dose-dependent increase in cells with &gt;2 centrosomes for each NRTI, but not for thymidine. In experiments designed to evaluate functional viability of amplified centrosomes, microtubule nucleation was arrested with nocodazole. When nocodazole was removed the cells with &gt;2 centrosomes had the ability to form spindle poles, and complete normal cell division. Overall the study shows that centrosome amplification occurs as a result of exposure to all NRTIs, and that cells with centrosome amplification are able to accomplish cell division. The resulting genomic instability and replication of cells with multipolar mitotic figures may contribute to the carcinogenic properties of the NRTIs. Amifostine is a strong antioxidant that selectively protects normal tissues against ionizing radiation damage and chemotherapeutic drug cytotoxicity. We hypothesized that amifostine might protect host organs from mitochondrial toxicity induced by NRTI therapy, and we first exposed human HIV-1-infected phytohemagglutinin (PHA)-stimulated T-cell blasts with WR1065 to observe any alteration in the antiretroviral capacity of AZT. Unexpectedly we found that WR1065 has antiretroviral properties [Patent NIH332.001VPC Organic Thiophosphate Antiretroviral Agents;Walker, D.M. et al., Environmental and Molecular Mutagenesis 50:460-472, 2009.]. Subsequent extensive WR1065 dosimetry studies in human T-cell blasts and in T-cell blasts taken from macaques infected long-term with Simian Immunodeficiency virus (SIV) showed substantial inhibition of both HIV-1 and SIV replication. HIV-1 replication (p24) in human T-cell blasts was inhibited by 50% at 13 M WR1065, a dose giving 80% cell viability. Additive antiviral activity was observed with the combination of AZT plus WR1065. Cultured T-cell blasts from macaques (n=3) chronically infected (14 months) with SIV were incubated 17 days with WR1065. Assays for viral replication (p27) and cell viability (MTS assay) revealed 100% inhibition of SIV replication, and 70% cell survival, in cells from all 3 monkeys cultured for 17 days. Therefore, WR1065 and its therapeutic precursor amifostine, may be useful, given either alone or in combination with NRTIs, as adjuvant antiretroviral therapy. Studies in animal models We are performing ongoing investigations of mitochondrial toxicity in fetal Erythrocebus patas monkeys, taken at birth, 1 year and 3 years of age, from dams exposed to human-equivalent protocols containing 3TC, AZT, AZT/3TC, AZT/ddI, 3TC/d4T, AZT/3TC/ABC or AZT/3TC/NVP. To model human clinical use, pregnant patas dams (n=3-4/group) are dosed during the last half of gestation, and the neonates for 6 weeks after birth. Biomarkers include: clinical chemistry;electron microscopy (EM) for mitochondrial morphology;scoring of damage in EM photos;quantitation of mitochondrial (mt)DNA using PCR-based hybrid capture-chemiluminescence immunoassay (HC-CIA);oxidative phosphorylation (OXPHOS) enzyme assays;and OXPHOS histochemical staining. Studies that include 1 or 2 NRTIs, with time points at birth and 1 year are complete, and published data for the heart and skeletal muscle have documented substantial mitochondrial morphological abnormalities and alterations of mtDNA levels that did not return to normal levels by 1 year of age. Brain mitochondria from NRTI-exposed patas, both at birth and 1 year of age, showed significant (p&lt;0.05) morphological damage by EM, whereas the liver EM photomicrographs showed some abnormalities but no significant difference from the controls. Brain and liver mtDNA, measured by HC-CA, were not significantly depleted at birth, but were depleted (p&lt;0.05) in all NRTI-exposed groups at 1 year of age. In brains and livers of 1 year old infants exposed in utero to the 2-NRTI combinations, mtDNA depletion was 41-52% and 37-56%, respectively. The data suggest that NRTI-exposed HIV-1-uninfected human infants may sustain persistent and possibly progressive brain and liver mitochondrial compromise as a result of in utero exposure to NRTIs.